2.1. Reagents

The DMDOSA (extractant N,N-dimethyl-N′,N′-dioctyl succinylamide) and DMDDSA (N,N-dimethyl-N′,N′- didecylsuccinylamide) were obtained by reaction of N-methyl octylamine and N-methyl decylamine with succinyl chloride ^[19], respectively. Both DMDOSA and DMDDSA were purified by distillation under vacuum. Structures of the extractants were shown in Fig. 1. The final products were characterized by elemental analysis, IR and ¹HNMR measurements, and the purity was over 98%. The corresponding data were given as follows:

Fig. 1Full-screen View

The structures of extractants. (a) N,N-dimethyl-N′,N′-dioctyl succinylamide (DMDOSA), (b) N,N-dimethyl- N′,N′-didecylsuccinylamide (DMDDSA)

DMDOSA: (yield 77.4%), pale yellow oil, 98% purity, b.p. 240 °C–242 °C/2–4 mmHg; ν_C=O(film)/cm^-1 1648.2; δ_H(300 MHz, DMSO), 0.857 (6H, t, C-CH₃), 1.239–1.260(20H, m, -CH₂-), 1.346–1.546(4H, m, -CH₂-C-N-), 2.764–2.940 (6H, t, N-CH₃), 3.200–3.292 (4H, m, N-CH₂-),2.481 (4H, t, OC-CH₂-CH₂-CO). Element analysis for C₂₂H₄₄N₂O₂ (Calcd.): C: 71.46% (71.69%), H: 11.94% (12.03%), N: 7.51% (7.60%).

DMDDSA: (yield 58.4%), white solid, 98% purity, b.p. 246 °C-248°C/2~4mmHg; ν_C=O(film)/cm^-1 1648.3; δ_H(300MHz, DMSO), 0.860(6H, t, C-CH₃), 1.239-1.260(24H, m, -CH₂-), 1.346-1.546 (4H, m, -CH₂-C-N-), 2.764, 2.939(6H, t, N-CH₃), 3.200-3.292(4H, m, N-CH₂-), 2.481(4H, m, OC-CH₂-CH₂-CO). Element analysis for C₂₆H₅₂N₂O₂ (Calcd.): C 73.53% (73.29%),H 12.34%(12.23%), N 6.60%(6.74%)。

Other agents employed in this work were of A.R. grades.

2.2. Extraction procedure

Equal volumes of aqueous phase and organic phase were vigorously mixed for 30 min to reach equilibrium at a constant temperature. The two phases were centrifuged and assayed by taking 100-500 μL aqueous phase. The U(VI) concentration in a sample was determined by Arsenazo-III visible spectrophotometric analysis, and the U(VI) concentration in organic phase was calculated by subtracting the aqueous concentrations from the total initial aqueous concentration of U(VI). The distribution ratio (D) was calculated by dividing the U(VI) concentration in the organic phase over the U(VI) concentration in aqueous phase. The D values reported were averaged from at least duplicate experiments, with errors of less than 5%.

Back extraction was conducted as follows: 2 mL of above organic phase after extraction and 2 mL pure water or 0.10 mol·dm^-3 Na₂CO₃ solution were vigorously mixed for 30 min. The two phases were centrifuged and assayed by taking 100 μL aqueous phase. The concentrations of U(VI) in the sample were determined by Arsenazo-III visible spectrophotometric analysis.

2.3. FT-IR characterization

The extractant solutions were shaken with a concentrated solution of UO₂(NO₃)₂, and centrifuged to collect the organic phase. The organic solvents were removed by evaporation and infrared spectra of the residuals were recorded in the wavenumber range of 4000-400 cm^-1 on a Perkin-Elmer Spectrum One FT-IR spectrometer, using KBr sample windows, in 60 scan times per sample, at a resolution of 2 cm^-1.

3.1. Effect of nitric acid concentration

The extraction of U(VI) with 0.20 mol·dm^–3 DMDOSA and DMDDSA (in cyclohexane) from nitric acid solutions of 1.00-5.00 mol·dm^-3 HNO₃ at 298 K were investigated.The results are shown in Fig.2. The distribution ratio of U(VI) increases with the HNO₃ concentration. This is similar to the results in HNO₃ medium systems ^[19], and can mainly be attributed to common ion effect of nitrate ion. Comparing the data of extraction of U(VI) with DMDOSA in toluene^[19] and in cyclohexane, one knows that it is beneficial to use cyclohexane as diluents at high nitric acid concentrations of over 4.0 mol·dm^-3. Another advantage of using cyclohexane as diluents, over toluene, is that cyclohexane as paraffin is environment friendly.

Fig. 2Full-screen View

Effect of HNO₃ concentration on U(VI) extraction.

3.2. Effect of extractant concentration

The extraction equilibrium of UO₂ ²⁺ from HNO₃ solution by DMDDDSA can be described as follows:

The conditional extraction constant is

where $C_{{\text{UO}}_{\text{2}}^{\text{2}+}}$ is the total concentration of UO₂²⁺ in the aqueous phase, $CUO2(NO3)2\cdot nDMDDSA$ is the concentration of the extracted species in the organic phase. Eq. (4) can be rearranged into Eq. (5) after taking the logarithm:

Where C_NO3^- is kept constant, and DMDDSA is in a large excess to UO₂²⁺ ions under the studied conditions, then the concentration of DMDDSA can be regarded as constant. Thus, the stoichiometric coefficients n can be determined by slope analysis of the relationship between logarithm distribution ratio and logarithm equilibrium concentration. Fig. 3 shows the log D as a function of logarithm DMDDSA and DMDOSA concentration.

Fig. 3Full-screen View

Effect of extractant concentration on U(VI) extraction.

The distribution ratio increases linearly with the DMDDSA or DMDOSA concentration, with the slopes of 2.17 for logD vs. logC_DMDDSA(O) and 2.25 for logD vs. logC_DMDOSA(O). These show that two molecules of amide are associated with one UO₂²⁺ ions in the extractable complex. However, the average number of TOSA ^[20] and TBSA ^[14] coordinated to U(VI) ion was estimated to be one. This is probably connected with nitric acid concentration or the diluents that have a major impact on the extraction behavior due to their polarity ^[21]. We propose that the difference of the coordinated extractant numbers is possibly due to the size of the substitute. The methyl group of DMDOSA or DMDDSA is smaller compared with octyl group of TOSA or butyl group of TBSA. The methyl group has less steric hindrance resulting in more extractant molecules into the coordination sphere. The extraction mechanism can be described as:

The stoichiometry of extracted species is the same as that obtained in toluene system ^[19], which shows that diluent does not affect the composition of extracted species.

FT-IR spectra of the organic phase with the extracted species were recorded (Fig. 4). Analysis of the spectra of extracted species was carried out in the C=O stretching region (1550-1760 cm^-1) using a curve fitting procedure. The broad band of C=O consists of two sub-bands and the peaks are centered at 1586 and 1648 cm^-1. Compared with the FT-IR spectra of the free extractants, the carbonyl group (C=O) stretching vibration of the extracted species shifts from about 1648 to 1586 cm^–1, which implies that the amide ligand directly coordinates to the uranyl ion through its carbonyl group. The appearance of bands at about 1030, 1281 and 1522 cm^-1 can be indexed to the coordinated nitrate ions.^[22,23] An additional band described as the U-O stretching vibration of the uranyl ion is observed at 934 cm^–1.

Fig. 4Full-screen View

IR spectra of the extractant loaded uranyl nitrate

3.3. Temperature effect on extraction

Temperature effect on extraction equilibrium of U(VI) was studied at 15 °C-55 °C (Fig.5). The distribution ratio decreases with increasing temperature. From the Van’t Hoff equations ^[24], the thermodynamic parameters can be calculated by:

Fig. 5Full-screen View

Effect of experimental temperature on the extraction of U(VI)

where R is the gas constant and C is a conditional constant. In this case, the slope in Fig. 5 is ΔH / (2.303R). From Eq. (8) and Fig. 5, we obtained ΔH = -26.33 ± 1.56 kJ·mol^-1 for DMDOSA and ΔH = -23.60 ± 0.95 kJ·mol^-1 for DMDDSA. In toluene, we got -23.9±1.7 kJ·mol^-1 for DMDOSA^[19]. The results indicate that cyclohexane is a better diluent. The change in enthalpy demonstrates that the extraction reaction is an exothermic process and it is disadvantageous for the extraction reaction at high temperature.

3.4. Effect of extractant structure

Comparing to the extraction of U(VI) with N,N, N′,N′-tetraalkylsuccinylamide ^[14,25], the adoption of unsymmetrical substituted alkyls, especially with methyl and straight long chain alkyl, does improve the extractability of succinylamide towards U(VI). The two straight long chain alkyls adjacent to nitrogen atom in succinylamide extractants lead to a big steric hindrance effect, which is possibly responsible for the difference of the formed extracted species. One alkyl group of DMDOSA or DMDDSA adjacent to nitrogen atom is a small group-methyl and the other is a bigger one-octyl or decyl, the stereochemical hindrance around the carbonyl oxygen atoms is kept to a minimum and both oxygen atoms can interact with uranyl ion ^[26]. On the other hand, the hydrophobic character of the ligands is attained in the presence of a longer alkyl-octyl or decyl. The intercept of log D vs. log D_amide (Fig. 3) was log K_ex, as given in Table 1.

Generally, the basicity, i.e. charge distribution of coordination group is related to the coordination ability. In view of atoms bonding with U(VI) are oxygen atoms, the charge distributions of carbonyl oxygen of DMDOSA and DMDDSA were calculated by employing density functional theory (DFT) with the B3LYP/6-31+G(d) level of the Gaussian 03program ^[27]. The results are given in Table 1.

It is obvious that log K_ex of DMDOSA is nearly equal to logK_ex of DMDDSA, which is possibly that the steric hindrance effect of octyl is not much different from that of decyl. Moreover, the charge distribution of carbonyl oxygen of DMDOSA is nearly equal to that of DMDDSA, hence the similar extraction result. Table 1 also shows that the distal carbon atoms do not influence the charge distribution of carbonyl oxygen effectively.

3.5. Back extraction

The results of stripping U(VI) loaded in the organic phases show that the back extraction of U(VI) extracted by DMDOSA or DMDDSA is efficient. The stripping efficiency is over 96% with pure water for one step, and nearly 100% with 0.10 mol·dm^-3 Na₂CO₃ solution.